ICU Bloodstream Infection Prediction: A Transformer-Based Approach for EHR Analysis

TL;DR

RatchetEHR is a transformer-based framework that improves ICU bloodstream infection prediction by effectively analyzing EHR data with a novel GCT component, outperforming traditional models.

Contribution

Introduces RatchetEHR, a transformer-based model with GCT for enhanced EHR analysis and BSI prediction in ICU settings, demonstrating superior performance over existing methods.

Findings

Outperforms RNN, LSTM, and XGBoost in BSI prediction

GCT component improves detection of structural relationships in EHR data

Effective even with small, imbalanced datasets

Abstract

We introduce RatchetEHR, a novel transformer-based framework designed for the predictive analysis of electronic health records (EHR) data in intensive care unit (ICU) settings, with a specific focus on bloodstream infection (BSI) prediction. Leveraging the MIMIC-IV dataset, RatchetEHR demonstrates superior predictive performance compared to other methods, including RNN, LSTM, and XGBoost, particularly due to its advanced handling of sequential and temporal EHR data. A key innovation in RatchetEHR is the integration of the Graph Convolutional Transformer (GCT) component, which significantly enhances the ability to identify hidden structural relationships within EHR data, resulting in more accurate clinical predictions. Through SHAP value analysis, we provide insights into influential features for BSI prediction. RatchetEHR integrates multiple advancements in deep learning which together provide accurate predictions even with a relatively small sample size and highly imbalanced dataset. This study contributes to medical informatics by showcasing the application of advanced AI techniques in healthcare and sets a foundation for further research to optimize these capabilities in EHR data analysis.